Refuseless API
Hosted abliterated GLM and DeepSeek models with OpenAI and Anthropic compatibility.
Good fit if you want a hosted abliterated model and the option to fund your balance by card. The API uses prepaid, per-token billing. Use the provider’s model list to confirm the ID accepted by your account.
Pricing and usage
| Model | Input | Output | Cached input | Context |
|---|---|---|---|---|
| DeepSeek V4.1 Flash 1017 Abliterated | $0.12 | $0.45 | $0.025 | 1M |
| GLM 5.3 Flash Abliterated | $0.12 | $0.50 | $0.025 | 1M |
| GLM 5.3 Abliterated | $1.00 | $2.00 | $0.15 | 1M |
As an example, 1 million input tokens and 250,000 output tokens on DeepSeek V4.1 Flash 1017 Abliterated would cost $0.233 at the listed rates, rounded to the nearest $0.001. This estimate excludes cached pricing and credits.
Getting started
- Sign up through the referral link and create an API key.
- Choose an abliterated model from the current catalog.
- Point a compatible client at the API base URL. Use the official quickstart for request format and authentication.
Base URLhttps://api.refuseless.com/v1
Content rules
Refuseless publishes general acceptable-use terms. Adult-content permission for specific endpoints is unverified.
Provider documentation reviewed. We have yet to run hands-on endpoint tests.
Docs and sources
Referral signup and a first deposit of at least $10.